Dan Semin
commited on
Commit
•
0470cf4
1
Parent(s):
b831448
update model card README.md
Browse files
README.md
CHANGED
@@ -22,7 +22,7 @@ model-index:
|
|
22 |
metrics:
|
23 |
- name: Accuracy
|
24 |
type: accuracy
|
25 |
-
value: 0.
|
26 |
---
|
27 |
|
28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -32,8 +32,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
32 |
|
33 |
This model is a fine-tuned version of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) on the xnli dataset.
|
34 |
It achieves the following results on the evaluation set:
|
35 |
-
- Loss:
|
36 |
-
- Accuracy: 0.
|
37 |
|
38 |
## Model description
|
39 |
|
@@ -53,90 +53,24 @@ More information needed
|
|
53 |
|
54 |
The following hyperparameters were used during training:
|
55 |
- learning_rate: 5e-05
|
56 |
-
- train_batch_size:
|
57 |
-
- eval_batch_size:
|
58 |
- seed: 42
|
59 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
60 |
- lr_scheduler_type: linear
|
61 |
-
- num_epochs:
|
62 |
|
63 |
### Training results
|
64 |
|
65 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
66 |
|:-------------:|:-----:|:-----:|:---------------:|:--------:|
|
67 |
-
| 0.
|
68 |
-
| 0.
|
69 |
-
| 0.
|
70 |
-
| 0.
|
71 |
-
| 0.
|
72 |
-
| 0.
|
73 |
-
| 0.
|
74 |
-
| 0.6924 | 0.33 | 8000 | 0.6793 | 0.7357 |
|
75 |
-
| 0.6807 | 0.37 | 9000 | 0.6604 | 0.7502 |
|
76 |
-
| 0.6636 | 0.41 | 10000 | 0.6309 | 0.7430 |
|
77 |
-
| 0.6616 | 0.45 | 11000 | 0.6039 | 0.7490 |
|
78 |
-
| 0.6561 | 0.49 | 12000 | 0.6051 | 0.7610 |
|
79 |
-
| 0.6545 | 0.53 | 13000 | 0.6354 | 0.7454 |
|
80 |
-
| 0.644 | 0.57 | 14000 | 0.6064 | 0.7466 |
|
81 |
-
| 0.6446 | 0.61 | 15000 | 0.6052 | 0.7554 |
|
82 |
-
| 0.6414 | 0.65 | 16000 | 0.6365 | 0.7422 |
|
83 |
-
| 0.6311 | 0.69 | 17000 | 0.6118 | 0.7546 |
|
84 |
-
| 0.6187 | 0.73 | 18000 | 0.5973 | 0.7538 |
|
85 |
-
| 0.619 | 0.77 | 19000 | 0.5863 | 0.7570 |
|
86 |
-
| 0.6108 | 0.81 | 20000 | 0.6212 | 0.7490 |
|
87 |
-
| 0.6136 | 0.86 | 21000 | 0.5810 | 0.7695 |
|
88 |
-
| 0.6018 | 0.9 | 22000 | 0.5799 | 0.7731 |
|
89 |
-
| 0.6198 | 0.94 | 23000 | 0.5548 | 0.7723 |
|
90 |
-
| 0.6047 | 0.98 | 24000 | 0.5964 | 0.7622 |
|
91 |
-
| 0.5636 | 1.02 | 25000 | 0.5805 | 0.7851 |
|
92 |
-
| 0.5267 | 1.06 | 26000 | 0.5540 | 0.7795 |
|
93 |
-
| 0.5067 | 1.1 | 27000 | 0.5388 | 0.7855 |
|
94 |
-
| 0.5304 | 1.14 | 28000 | 0.5482 | 0.7799 |
|
95 |
-
| 0.5332 | 1.18 | 29000 | 0.5290 | 0.7859 |
|
96 |
-
| 0.5154 | 1.22 | 30000 | 0.5475 | 0.7799 |
|
97 |
-
| 0.524 | 1.26 | 31000 | 0.5305 | 0.7900 |
|
98 |
-
| 0.5236 | 1.3 | 32000 | 0.5691 | 0.7871 |
|
99 |
-
| 0.5154 | 1.34 | 33000 | 0.5642 | 0.7739 |
|
100 |
-
| 0.5248 | 1.39 | 34000 | 0.5590 | 0.7643 |
|
101 |
-
| 0.5077 | 1.43 | 35000 | 0.6064 | 0.7715 |
|
102 |
-
| 0.5147 | 1.47 | 36000 | 0.5343 | 0.7948 |
|
103 |
-
| 0.5041 | 1.51 | 37000 | 0.5375 | 0.7867 |
|
104 |
-
| 0.5054 | 1.55 | 38000 | 0.5660 | 0.7727 |
|
105 |
-
| 0.5053 | 1.59 | 39000 | 0.5479 | 0.7859 |
|
106 |
-
| 0.5009 | 1.63 | 40000 | 0.5080 | 0.7960 |
|
107 |
-
| 0.5081 | 1.67 | 41000 | 0.5139 | 0.7920 |
|
108 |
-
| 0.5013 | 1.71 | 42000 | 0.5385 | 0.7904 |
|
109 |
-
| 0.4972 | 1.75 | 43000 | 0.5257 | 0.7928 |
|
110 |
-
| 0.4987 | 1.79 | 44000 | 0.5056 | 0.8020 |
|
111 |
-
| 0.4863 | 1.83 | 45000 | 0.5030 | 0.8004 |
|
112 |
-
| 0.5 | 1.87 | 46000 | 0.5157 | 0.7980 |
|
113 |
-
| 0.4926 | 1.91 | 47000 | 0.5505 | 0.7924 |
|
114 |
-
| 0.4893 | 1.96 | 48000 | 0.5286 | 0.8004 |
|
115 |
-
| 0.4755 | 2.0 | 49000 | 0.5216 | 0.8036 |
|
116 |
-
| 0.3855 | 2.04 | 50000 | 0.6087 | 0.7884 |
|
117 |
-
| 0.3731 | 2.08 | 51000 | 0.5485 | 0.8064 |
|
118 |
-
| 0.3698 | 2.12 | 52000 | 0.5398 | 0.8080 |
|
119 |
-
| 0.3702 | 2.16 | 53000 | 0.5454 | 0.8 |
|
120 |
-
| 0.3688 | 2.2 | 54000 | 0.5512 | 0.8068 |
|
121 |
-
| 0.3683 | 2.24 | 55000 | 0.5423 | 0.8060 |
|
122 |
-
| 0.3704 | 2.28 | 56000 | 0.5383 | 0.8084 |
|
123 |
-
| 0.3758 | 2.32 | 57000 | 0.5398 | 0.8161 |
|
124 |
-
| 0.3781 | 2.36 | 58000 | 0.5481 | 0.8088 |
|
125 |
-
| 0.3697 | 2.4 | 59000 | 0.5465 | 0.8056 |
|
126 |
-
| 0.3706 | 2.44 | 60000 | 0.5488 | 0.7988 |
|
127 |
-
| 0.3704 | 2.49 | 61000 | 0.5395 | 0.8052 |
|
128 |
-
| 0.3648 | 2.53 | 62000 | 0.5463 | 0.8068 |
|
129 |
-
| 0.36 | 2.57 | 63000 | 0.5400 | 0.8052 |
|
130 |
-
| 0.3661 | 2.61 | 64000 | 0.5542 | 0.8068 |
|
131 |
-
| 0.3555 | 2.65 | 65000 | 0.5424 | 0.8044 |
|
132 |
-
| 0.3551 | 2.69 | 66000 | 0.5269 | 0.8124 |
|
133 |
-
| 0.3608 | 2.73 | 67000 | 0.5382 | 0.8129 |
|
134 |
-
| 0.35 | 2.77 | 68000 | 0.5461 | 0.8108 |
|
135 |
-
| 0.3457 | 2.81 | 69000 | 0.5477 | 0.8084 |
|
136 |
-
| 0.3516 | 2.85 | 70000 | 0.5345 | 0.8104 |
|
137 |
-
| 0.3499 | 2.89 | 71000 | 0.5344 | 0.8129 |
|
138 |
-
| 0.3513 | 2.93 | 72000 | 0.5279 | 0.8120 |
|
139 |
-
| 0.3442 | 2.97 | 73000 | 0.5306 | 0.8108 |
|
140 |
|
141 |
|
142 |
### Framework versions
|
|
|
22 |
metrics:
|
23 |
- name: Accuracy
|
24 |
type: accuracy
|
25 |
+
value: 0.8076305220883534
|
26 |
---
|
27 |
|
28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
32 |
|
33 |
This model is a fine-tuned version of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) on the xnli dataset.
|
34 |
It achieves the following results on the evaluation set:
|
35 |
+
- Loss: 1.0082
|
36 |
+
- Accuracy: 0.8076
|
37 |
|
38 |
## Model description
|
39 |
|
|
|
53 |
|
54 |
The following hyperparameters were used during training:
|
55 |
- learning_rate: 5e-05
|
56 |
+
- train_batch_size: 128
|
57 |
+
- eval_batch_size: 128
|
58 |
- seed: 42
|
59 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
60 |
- lr_scheduler_type: linear
|
61 |
+
- num_epochs: 7
|
62 |
|
63 |
### Training results
|
64 |
|
65 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
66 |
|:-------------:|:-----:|:-----:|:---------------:|:--------:|
|
67 |
+
| 0.3328 | 1.0 | 3068 | 0.5433 | 0.8036 |
|
68 |
+
| 0.259 | 2.0 | 6136 | 0.5708 | 0.8008 |
|
69 |
+
| 0.2023 | 3.0 | 9204 | 0.6475 | 0.8048 |
|
70 |
+
| 0.1362 | 4.0 | 12272 | 0.7661 | 0.7972 |
|
71 |
+
| 0.0945 | 5.0 | 15340 | 0.8333 | 0.8008 |
|
72 |
+
| 0.0665 | 6.0 | 18408 | 0.9312 | 0.8092 |
|
73 |
+
| 0.0463 | 7.0 | 21476 | 1.0082 | 0.8076 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
74 |
|
75 |
|
76 |
### Framework versions
|